How Artificial Intelligence Is Making The Eyecare Profession Smarter
We’re seeing signs everywhere of the many ways artificial intelligence (AI) is advancing healthcare. Amazon is working with a major medical center to deploy AI applications. Cleveland Clinic has launched a center to advance AI in healthcare. In April, Google, Fitbit, and other tech giants, associations, and other health industry organizations came together to provide best practices for application of AI in healthcare. And, more than a year ago, the FDA approved the first autonomous AI system for detecting diabetic retinopathy in images taken by retinal cameras.
So, what exactly is AI? Broadly defined, it’s a set of technologies that use algorithms to mimic neural functions to perform complex analytic tasks that range from visual perception and speech recognition to decision making.
Today, AI is everywhere, making its way into nearly every area of our daily lives—in the form of virtual assistants like Amazon’s Alexa, even apps that provide analytics that predict when you’re likely to cheat on your diet. On a larger scale, researchers at Google’s DeepMind are demonstrating the power of AI for predicting everything from wind energy production to eye disease.
In fact, nearly every industry is feeling the impact of AI. The financial industry is employing AI for fraud protection, portfolio management, and more. It’s helping manufacturers improve productivity, efficiency, and safety. It’s also setting new standards in transportation, media and entertainment, retail, and more.
AI and Healthcare
When it comes to healthcare, AI is creating a paradigm shift. In Japan, for example, an AI-powered system built with Google’s TensorFlow architecture is helping radiologists detect colon cancer in less than a second. In fact, with its exceptional ability to analyze massive amounts of data, AI holds great promise for helping healthcare professionals detect and diagnose diseases like cancer at early stages—and soon, for providing valuable recommendations for treatment approaches.
Many studies have already proven that AI has the potential to be a valuable tool for radiology and pathology, two specialties that rely heavily on medical imaging for diagnoses. And, precisely because of its extraordinary image and pattern recognition capabilities, AI is quickly growing in popularity among eyecare professionals.
AI technology makes it possible to develop algorithms that can be used to identify the characteristics of healthy eyes and those affected by various conditions. This is done by “feeding” the software a database of images that are analyzed incrementally in order to begin recognizing relationships and patterns to classify images and provide results with highly accurate sensitivity and specificity.
AI and Diabetic Retinopathy
Diabetic retinopathy (DR) is the first frontier where AI is dramatically impacting eyecare. That’s significant, because early diagnosis of DR is important. DR, a condition that damages the blood vessels in the retina, is a leading cause of blindness. Every year, more than 24,000 people lose their sight to DR from severe complications of diabetes. According to the CDC, the number of Americans with the condition is projected to grow from 7.7 million in 2010 to 14.6 million 2050. The World Health Organization (WHO) has stated that 80% of all visual impairment caused by DR can be cured with regular screening and early detection.
When diagnosed early, DR can usually be managed and the worst outcomes prevented. Yet, because there are no symptoms in its early stages, it is difficult to detect. AI provides the capability to identify microscopic lesions on the eye to enable early diagnosis of the disease.
Now, telehealth, combined with AI technology and fundus imaging, offers eyecare providers a cost-effective and powerful solution that can expand access to care, while providing earlier detection of DR and prevention of unnecessary vision loss. With advances in technology and the availability of large data sets of retinal images, AI systems for detecting DR, as well as other common eye diseases such as age-related macular degeneration and glaucoma, are now available.
20/20NOW’s Patented AI Technology
20/20NOW, the inventor and leader in ocular telehealth, has incorporated the use of AI in our comprehensive eye exam. Our ocular telehealth solution integrates FDA-approved AI with images produced by a wide-field non-mydriatic retinal camera. Using AI, the retinal images captured by the cameras are immediately enhanced to identify lesions in the retina, as well as other conditions typically undetectable to the human eye.
This technology, which 20/20NOW calls Eyelogic™, assists doctors in diagnosing critical eye diseases like DR in their early stages, a critical step toward preventing blindness and lowering healthcare costs. Our patented EyeLogic AI technology uses an algorithm built from analysis of more than 75,000 fundus images found in the Eye Care Picture Archive and Communication System (ecPACS) telemedicine database. The study, published in a recent issue of Ophthalmology, concluded that:
“a fully data-driven artificial intelligence-based grading algorithm can be used to screen fundus photographs obtained from diabetic patients and to identify, with high reliability, which cases should be referred…for further evaluation and treatment. The implementation of such an algorithm on a global basis could reduce drastically the rate of vision loss attributed to DR.”
See Eyelogic for Yourself
20/20NOW’s autonomous AI diagnostic system uses an algorithm based on the way that clinicians evaluate DR—using machine learning to detect the exudates, hemorrhages, micro-aneurysms, and other lesions that can indicate a DR diagnosis—and provides a standard of care equivalent to an in-person exam.
This set of images represents a diagnosis of diabetic retinopathy with a conventional image on the left and an image enhanced by the 20/20NOW Eyelogic AI program on the right. The circles demonstrate areas of low, medium and high risks of dot hemorrhages and exudates.
The second set of images represents a diagnosis of dry age-related macular degeneration demonstrating focal areas of concern of hard exudates and drusen. A conventional image is on the left and an image enhanced by Eyelogic AI is on the right.
If you’d like to know more about how our patented Eyelogic intelligence uses AI to detect eye diseases early, when they’re most treatable, visit http://for2020now.com. Even better, contact us to arrange an in-person demonstration by visiting the Contact Us page at our website and click “Schedule a demonstration.” Or call us at 844-843-2020.”Back to Blogs